© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 89284
    Prev
    Next
    INDEX
    Explanations

    Code/Programming

    np_max-act · gemini-2.0-flash

    The neuron activates on code‐style “declarations” or “prolog” lines (e.g. XML version tags, RewriteEngine directives, and similar markup/config headers).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_11/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.11.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    眼
    -0.07
     femme
    -0.07
     AHL
    -0.06
    asyonu
    -0.06
    oca
    -0.06
    ولد
    -0.06
    ilerini
    -0.06
     adaptations
    -0.06
     lymph
    -0.06
     rights
    -0.06
    POSITIVE LOGITS
    nz
    0.07
     isOpen
    0.07
    Preferred
    0.06
    deleted
    0.06
     '''
↵
    0.06
    ै,
    0.06
    .Ship
    0.06
     {*}
    0.06
     bush
    0.06
    >%
    0.06
    Activations Density 0.069%

    No Known Activations